Reevaluating Coercion: Paternalism And Less-Than-Voluntary Interventions In Public Health – Bjorn Watsjold

Public health interventions, with few exceptions, are designed and intended to persuade individuals to willfully change behavior. Interventions promoting weight loss have used education and advertising to ensure that individuals are aware of increasing rates of obesity (1) and the increased health risks obesity presents (2). Yet for many individuals, knowledge of risks and the intent to improve health do not easily manifest behavioral change: individuals may not believe they have the willpower to change, or they may not have adequate means to achieve a desired behavior. Dean Karlan, a Yale economics professor, founded StickK.com to improve on the failures of commercial weight loss programs. Studies have shown that after initial successes with commercial programs, most people fail to maintain their target weight after six months, and most will have returned to their baseline weight after one year (3). To aid in the achievement and maintenance of health goals, StickK offers a “commitment contract” that improves upon traditional behavior-model interventions with “two well-known principles of behavioral economics: people don’t always do what they claim they want to do, and incentives get people to do things (4).”

The commitment contract offered by StickK has been developed to encourage and enforce weight-loss programs, though it may be used to reach any health goal. Users register with the site, identify a personal health goal (e.g. weight loss), a weekly goal (lose two pounds per week until my target weight is reached), and then choose a level of monetary incentive to maintain their willpower ($50 per week). If users are successful, they keep the incentive money, and if not, they must donate it to charity. As an additional incentive, users are encouraged to choose a charity to which they are personally opposed (4). With the addition of a monetary penalty, the commitment contract is intended to help users to reach and maintain health goals with which they have not had success due to limitations of willpower. StickK may be analyzed as a public health intervention by considering three components: the model of behavior used to design the intervention, the method intended to induce behavioral change, and the group targeted by the intervention.

Analysis Of Commitment Contracts Using Behavioral Models
Public health interventions traditionally utilize models of behavior to understand the ways people analyze information, choose to act, and then act upon their decisions (5). Intervention design, therefore, is an attempt to manipulate individuals at specific stages of decision making or to manipulate the stages themselves. The commitment contract uses elements of three models, the Health Belief Model (HBM), the Theory of Planned Behavior, and the Transtheoretical Model (5, 6).

StickK employs the HBM by focusing on the balance of incentives and penalties that incite behavioral change (6). The commitment contract manipulates the components of the HBM by redefining the susceptibility and severity of a problem as immediate, monetary costs. This may serve to increase substantially the perceived costs of inaction and also to personalize those costs: for many individuals, a weekly monetary penalty may be both more easily understood and sharply felt than a long-term health benefit or burden. Though change is incited through a rational benefit-burden analysis, the costs are clearly defined and calculated to be effectively severe. The health benefits of behavioral change therefore become secondary to the artificially imposed financial burden.

The commitment contract is assumed to impose sufficient perceived costs to overcome the dislocation between intention and behavioral change that is a central weakness of the HBM. Beyond manipulation of incentives, the contract is also intended to improve self-efficacy, part of the Theory of Planned Behavior, here defined as the continuous reevaluation of one’s successes and failures in achieving an intended change (5). StickK directly targets individuals who doubt their own willpower, and therefore lack self-efficacy necessary to achieve a behavioral change. Users transfer authority to an outside party to assess penalties for non-compliance, thus increasing users’ likelihood of achieving their goals. The commitment contract, however, uses a narrow definition of self-efficacy to the extent that the target group is supposed to have failed due to lack of willpower but not due to other potential barriers, such as lack of social support, inadequate methods, or physiological challenges that would prevent achieving the desired behavioral change. Failing to address internal or external barriers not under individual control limits severely the usefulness of the contract as an intervention by assuming that people already possess the means to achieve behavioral change.

The Transtheoretical Model categorizes potential users at the action or maintenance stages (5). Individuals who have not reached the preparation or action phases, or who lack adequate information or programs to achieve the behavioral change, will not only not be successful in enacting and maintaining change, but the added disincentives of the commitment contract may cause significant harm. As the contract may only be broken for medical reasons, it may enforce unhealthy or ineffective behaviors (e.g. anorexia) in unprepared users while continuing to penalize failures. As the program does not include education or methods, it may be better used to support existing effective methods that fail due to non-compliance, thus allowing users to achieve goals they have been previously unable to despite adequate preparation, means, or support.

The Failure Of Incentives
The commitment contract utilizes monetary incentives to rationalize health risk-benefit analyses for users. While there is considerable evidence to demonstrate the effectiveness of monetary incentives to improve compliance to behavioral change (7), this effect is limited relative to size—there is an upper limit to increased compliance that may be gained by offering compensation. In predicting this effect on subjects, the limit for incentives has been reached at relatively low levels, only $50 in one study of physicians (8). Based upon rational decision-making, one would assume that larger monetary incentives created greater impetus for behavioral change, and yet it appears that people are no more rational calculating monetary incentives than they are when calculating health benefits. Contrary to the economic proposition that “incentives get people to do things,” it may be the case that redefining incentives in monetary terms will not be more effective in overcoming barriers between intention and behavioral change.

Enforcement is a second weakness of the commitment contract, as the user must recruit friends or acquaintances to verify and report success and failures. While third parties sign contracts with StickK to act as referees, the use of monetary incentives and the potential financial loss attached to weekly failures creates a strong incentive to cheat, which would render the contract completely ineffective. While StickK may assume that users are motivated and honest, the conflict of interest created by attaching monetary incentives to health benefits, which is compounded by the disconnection between the referees and StickK itself, undermines the usefulness of the contract as an intervention.

Using behavioral model analysis, the weakness of the commitment contact as a public health intervention are apparent: without providing means, methods, or support for users, self-efficacy is improved only to the extent that users are more harmed by failure. The contract does not significantly improve upon or overcome the behavior models it applies, because the mechanism used, monetary incentives, are still subject to the limitations of rational decision-making. Without removing non-rational or external barriers, the commitment contract cannot improve upon a rationally-based model. StickK may only be an appropriate method for the presumably small contingent that designed it: it is a health program by economists, for economists.

Ulysses Commitments: From Mental Health To Public Health
There is a saga of The Odyssey in which Ulysses wishes to hear the Sirens’ song, but realizing that he would be unable to resist them and would be driven mad, he creates an agreement under which his present rational desires would be upheld even against his future will. He ties himself to the ship’s mast and requires his men to not untie him under any circumstances and without regard for his pleas, and thus to prevent him from jumping overboard to answer the Sirens’ call (9). A Ulysses commitment, a reference to this saga, is made by patients with mental illness to authorize future treatment while their illness is controlled. Should patients undergo radical behavioral change or fail to comply with their treatment regimen they may be forced to undergo treatment, remedying their desire for non-treatment (9). Prior to establishing the Ulysses commitment as a legally binding contract, mentally ill patients could not be legally forced to comply with treatment unless they presented a legitimate danger to themselves or others. The contract allows for them to be treated in the initial stages of non-compliance: rather than waiting for significant harm to occur, they are treated and returned to their baseline rationality while minimizing potential harm. Though inadequately applied by StickK, the Ulysses commitment presents a novel use of non-voluntary interventions in public health by allowing individuals to submit to outside control to reinforce desired behavioral changes. While there are inherent difficulties in the use of commitment contracts as public health, the basic elements of the commitment contract are not foreign to health care.

Health interventions, whether at the personal level, as undertaken by a physician and patient, or at a group or population level by a public institution and citizens, require the autonomy of the individual to be respected. Interventions that enforce behavioral change involuntarily are considered paternalistic, and may be considered a threat to individual rights or self-reliance (10). To be considered legal and ethical, paternalistic interventions must alleviate a greater harm to the individual or to others than the harm posed by abridgment of individual autonomy, yet paternalistic interventions do exist: municipal water is chlorinated and fluoridated, the sale of alcoholic beverages and cigarettes is controlled, and drivers are compelled to wear seatbelts when driving. Coercive interventions are intended to inculcate positive health behaviors, such that they are internalized and become voluntary (10). From a theoretical perspective, these measures remove intentions or subordinate them to behavior through the application of mandates, eliminating self-efficacy for those who feel unable to achieve positive behaviors at the expense of those who are forced to comply to mandated behaviors.

The Ulysses contract is both voluntary and non-voluntary, as it enforces the current desires and autonomy of an individual against the individual’s future desires. Yet for individuals who are consistently unable to achieve or maintain behavioral change, a contract may allow them to overcome their own foreseen failures. To be used in public health, Ulysses contracts would allow self-coercion or “self-paternalism” (11) to improve self-efficacy and to control future actions. Ulysses contracts may include interventions for non-autonomous individuals, such as those undergoing mandatory drug or alcohol rehabilitation. Involuntary commitment is possible for addicts who become potentially harmful, because the physical dependence of addiction is considered a loss of autonomy which may be remedied paternalistically—individuals suffering from addiction can only regain their autonomy if they are treated for addiction, to which, it is assumed, they will not voluntarily submit. Ulysses contracts would allow preventive consent to rehabilitation or commitment for individuals in the process of treatment, such that family members or people involved in social support may trigger commitment or rehabilitation before a patient fails, not once the damage of relapse has been suffered.

Broader use of Ulysses contracts in public health would require abridgment of autonomy for autonomous individuals, but with careful regulation Ulysses contracts may present a temporary loss of autonomy as a means to increase future autonomy, by enforcing behavioral change to improve if not also prolong life. Analysis of StickK’s commitment contract shows that self-paternalism will be most effective for individuals who have attempted a behavioral change and failed to maintain the behavioral change due to lack of willpower or self-efficacy and despite adequate means and support. But individuals who have failed to remedy or improve a significant health concern, such as smoking cessation or weight-loss for morbid obesity, may wish to enter a non-voluntary program until they are able to maintain the behavior voluntarily. As an intervention, the contract would require the provision of effective means and support to overcome other barriers to self-efficacy, but by creating programs for ethical, legal self-paternalism, we allow individuals to rationally choose behavioral change while overcoming non-rational and non-voluntary barriers to enacting and maintaining behavior (12). While commitment contracts have not yet been applied effectively as a public health intervention, with proper regulation and public understanding, the potential benefits of self-paternalism may give many individuals the means to improve their own self-efficacy and achieve behavioral change not previously possible.

REFERENCES
1. Centers for Disease Control and Prevention. Obesity Trends Among US Adults. Atlanta, GA: Center for Chronic Disease Prevention and Health Promotion, 2007.
2. World Health Organization. Global Strategy on Diet, Physical Activity and Health: Obesity and Overweight. Geneva: World health Organization, 2003.
3. Heshka S, et. al. Weight Loss With Self-Help Compared With a Structured Commercial Program: A Randomized Trial. Journal of the American Medical Association 2003. 289:1792-1798.
4. StickK. About StickK.com. New York: StickK. http://www.StickK.com/about.php#
5. Edberg M. Essentials of Health Behavior: Social and Behavioral Theory in Public Health. Sudbury, MA: Jones and Bartlett, 2007.
6. Salazar MK. Comparison of Four Behavioral Theories: A Literature Review. AAOHN Journal 1991; 3:92-94.
7. Moore, M. Interactive Counseling Reduces Risky Sexual Behavior and Infections. Family Planning Perspectives 1999; 4:202-203.
8. James, J.M., and Bolstein, R. Large Monetary Incentives and Their Effect on Mail Survey Response Rates. The Public Opinion Quarterly 1992; 4:442-453.
9. Puran, N. Ulysses Contracts: Bound to Treatment or Free to Choose? The York Scholar 2005; 2:42-51.
10. Cottam R. Is Public Health Coercive Health? The Lancet 2005. 366:1594.
11. Dresser, R. Bound to Treatment: The Ulysses Contract. The Hastings Center Report 1984; 3:13-16.
12. Jolls, C. Contracts as Bilateral Commitments: A New Perspective on Contract Modification. The Journal of Legal Studies 1997; 1:203-237.